Talk Image manipulation and processing using NumPy and SciPy

Abstract

This tutorial will show a bag of basic recipes in order to efficiently manipulate and process images in the form of NumPy arrays.

Target audience Scientists and engineers working with images
Prerequisites Being able to code Python scripts and use an interactive Python shell + some knowledge of NumPy
Software requirements IPython, NumPy, SciPy, Matplolib, and optionally scikits.learn and scikits.image

Topics covered

  • I/O: how to open different image formats, how to open raw images, how to deal with very large raw files.
  • Basic visualization of images, and interaction with image data
  • Transforming images: changing the size, resolution, orientation of an image; image filtering; image segmentation.
  • Extracting information from images: measuring properties of segmented objects; image classification
  • (Very brief) discussion on some advanced image processing Python modules (scikits.image, CellProfiler, ITK, ...)

This tutorial will by no means be a course on digital image processing. It is rather a bag of tricks on how to tinker with images, and how to use the goodies of Python/NumPy/SciPy to make this task easier. A large part of the talk will be devoted to hands-on exercises using the NumPy, SciPy and Matplotlib modules. Some other modules will be mentioned during the tutorial for more advanced uses.

tagged by
no related entity